Sarra Ben Abbès

Led European projects for developing domain-specific AI in electric vehicle charging and European common data spaces in transport and energy sectors. Led the knowledge graph architecture and semantic data modeling for electric vehicle charging networks, ensuring scalable and efficient data interoperability between charging point operators. Drove the semantic layer architecture and strategy for integrating knowledge graphs in large-scale projects related to energy and mobility. Led the cloudification of the entire semantic data management process on Google Cloud Platform (GCP), including the transformation of raw data into semantic data with Apache NiFi, and implemented storage solutions in GraphDB, ensuring scalable and efficient data handling and enhancing data retrieval, query capabilities and decision-making. Developed and maintained Python scripts for data processing, analysis, and automation of workflows, collaborating with cross-functional teams to integrate semantic data solutions into existing systems and applications. Ensured data integrity and consistency throughout the cloud migration and transformation processes. Monitored AI trends impacting Gireve’s data services. Integrated Gireve teams with past projects. Framed and supervised data scientists in developing short- and medium-term recharge forecasting algorithms. Contributed to research project portfolio in Data domain, directly engaging in setup and execution. Collaborated with universities, research centers, and other companies in applied AI within mobility domain like Telecom ParisTech, CEA, Fortum, etc. Organized knowledge-sharing sessions, presentations, and workshops for disseminating AI advancements. Contributed to knowledge dissemination through publications, conference presentations, and industry events. Influenced AI standardization efforts in electric mobility sector through research activities (e.g., semantic modeling of the electromobility domain). Writing of activity reports and documents related to R&D work. Authored a scientific paper on electromobility in the energy data space, contributing to industry knowledge and best practices. Skills: Data Governance · Data Interoperability · Artificial intelligence · Electromobility domain · Standards and Data Strategies · Research and Innovation · Data Management · Project Management

Expert in Explainable Artificial Intelligence, Linked Data, Natural Language Processing (NLP), Semantic Web, Machine Learning, and Deep Learning. Leveraged Web Semantics (ontologies and taxonomies) by integrating NLP, Deep Learning, Machine Learning, Multi-agent Systems, and Blockchain. Modeled complex domains (e.g., energy systems, building infrastructures) using ontologies and linked data, enabling inference and reasoning. Developed ontologies and knowledge graphs to enhance urban infrastructure management, supporting design and O&M phases (e.g., street lighting, energy contracts). Led the creation of a domain-specific ontology to improve probabilistic graphical models in renewable energy projects. Designed and managed large-scale knowledge graphs using technologies like Neptune, GraphDB, RDF(S), and SPARQL for scalable and efficient data handling. Implemented a knowledge graph visualization and exploration tool, applied in: - Optimized Urban Planning: Facilitated intuitive navigation through urban planning ontologies for architects and engineers. - BIM (Building Information Modeling): Enhanced exploration of ontologies and RDF graphs in both design and O&M phases. - Public Lighting Management: Improved decision-making by visualizing relationships between streetlights and associated components (e.g., electrical cabinets, networks). - Simplified SPARQL query generation and developed user-friendly interfaces for non- expert stakeholders. Identified business needs and implemented innovative POCs across domains (e.g., buildings, HVAC, IoT, energy). Directed projects in Agile methodology, addressing energy-related contexts (e.g., wind turbines, lighting, roads). Collaborated with academics (Marne-la-Vallée, Saint-Etienne, Epita, Paris Descartes, Dauphine) and supervised interns and PhD students. Published work in national and international conferences and contributed to workshops like DL4NLP (EGC), DeepOntoNLP (ESWC), and OnUCAI (KR). Played a key role in European projects H2020 Platoon (https://cordis.europa.eu/ project/id/872592) and Enershare (https://enershare.eu/), leading work-packages and coordinating initiatives.

Storyzy have developed (1) a search engine that finds news quotes from publisher's articles on any topic, and (2) a robot that is able to automatically identify every new fake news site. Storyzy's solutions are powered by artificial intelligence and natural language processing (NLP). ************************* Role at Storyzy ************************* * Implement innovative solutions to solve problems related to the semantic analysis of citations in press articles in real time. * Enrichment of the Storyzy citation search engine to improve service quality * Implement solutions to check the accuracy of quotes and detect fake news sites. * Project management: semantic classification of documents, co-reference resolution, inte- gration of deep learning models to solve NLP problems, mapping between several semantic resources, approach evaluation protocol, automatic text summarization, sentiment analysis, etc. * Extraction of information from text documents * Supervision of trainees (4 trainees) * Control and validation of the results delivered via data consistency and non-regression testing tools * Writing technical documents * Management: Master’s degree trainees Technologies and platforms: * Big Data Technologies: Spark, NoSQL Database (MongoDB), Akka, ElasticSearch, Hadoop, Spark, Machine Learning techniques, classification, clustering, recommendation systems. * Semantic Web Technologies: Protected, Onto-Map, Virtuoso, Jena, RDFlib, RDF(S), OWL, SKOS, SPARQL 1.1 federated query, ontologies. * NLP Technologies: NLTK, JateToolkit, Stanford CoreNLP, Nooj, Gate, Named Entity Recognition, Treetagger, dependency relationship extraction, information extraction, text generation * Programming languages: Java, Python * Other techniques: Framework Spring, Redis, RabbitMQ, Jenkins, Maven, SVN, Git

Modélisation et implémentation d'une application multitiers pour la gestion des assurances IDA ( Indemnisation Directe de l’Assuré). Compétences techniques : J2EE, EJB3, XML, UML Logiciels : Eclipse, JBOSS, PostgreSQL, Merise, PowerAMC

Développement d'un système de gestion de documents des affiliés en utilisant le système de gestion de contenu Alfresco Compétences techniques : Java, XML Logiciels : Eclipse et Alfresco.